L'Ipotesi del Mapping Lessicale e l'apprendimento dei verbi in italiano come lingua seconda con particolare attenzione ai "riflessivi".

This research paper aims to test the plausibility of the Lexical Mapping Hypothesis of Processability Theory for Italian L2 according to its recent formulation (Pienemann, Di Biase, Kawaguchi 2005; Bettoni, Di Biase in press). In particular, we will test the implicationality of the three stages of the Lexical Mapping Hypothesis, later focusing our attention on the third stage of syntactic development, the stage of non-default mapping (NDF). For this stage, the research will focus on the study of «si-constructions» that require non-canonical mapping. Starting from the description of «si-constructions» according to the three structures of sentence representation (argument structure, functional structure and constituent structure) set out by Lexicon Functional Grammar (Bresnan 2001; Dalrymple 2001; Falk 2001), we will then proceed upon their classification on the basis of the type of mapping required (canonical vs non-canonical). The study is based on a corpus of Italian L2 learners made up of 24 adult learners with various L1s. With this paper we will demonstrate how the acquisition of the «si-constructions» does not happen in a uniform and simultaneous manner, but follows a sequence of acquisition determined by factors that relate to the nature of the topics of the verbal predicate and their mapping onto grammatical functions, and that also affect the pragmatic level of such constructions.